Hacker News Reader

Comment by maegul | Hacker News Reader

In the abstract this is an excessive take.

The point is that trust is a major component of scientific work and how it functions collectively. A effect being that when that is violated a lot breaks down with a god amount of collateral damage.

With increasing complexity in research and pressure to produce and publish more, it’s a growing weakness.

Bottom line is that team work and trust are now indelible parts of research in a culture predicated on individual success and contributions.

Honestly not sure how science adjusts.

View on HN

I see it as a symptom of our culture where benefits flow uphill and blame flows downhill.

When writing papers and presenting at conferences, the PI was all about how he was an integral part of the research and if you asked why he was presenting instead of the post-doc he would have talked about his responsibilities to supervise and guide.

Once the fraud came out, he is just a poor victim who can’t be expected to know the post-doc wasn’t even using the equipment and was making up data completely.

This is how every enterprise works.

The CEO gives the talk about the awesome product because they set the roadmap.

Benefits flow uphill, blame flows downhill is literally how any hierarchical organization works.

Now, you might want to say that science shouldn't be hierarchical, but we don't have any flat organizations like this that involve the kinds of mentoring and training required to build new scientists.

Not in every hierarchical organization.

For example, a ship’s captain takes ultimate responsibility for any collision or sinking, whether or not they were actually steering the ship. There was a tradition of the captain being the last one to leave a sinking ship.

Many cultures had a tradition of executing commanders that suffered defeat.

I think our culture is an anomaly where blame does not flow uphill as well.

Science is based on trust. All incentices are against research that verifies results: you don’t get funding, you don’t get published, you are not doing original research etc. If we cannot trust the results, everything is a house of cards, because not only are the papers wrong, also all papers depending on that paper may be affected. The scientific system should not be based on blind trust but have processes in place that makes it more resilient.

I don't see how it's not clear with the reproducibility crisis that trust is misplaced and what we are doing isn't working.

Reply on news.ycombinator.com